Methods and Systems for Single Frame Autofocusing Based on Color-Multiplexed Illumination

1. A system to capture a whole slide image, the system comprising: a microscope system with a digital camera, configured to capture a digital image of a sample; and a computing system operatively coupled to the camera, the computing system programmed to: capture a bright field image of the sample; capture a digital image of the sample illuminated by a first wavelength light at a first incident angle and by a second wavelength light at a second incident angle; determine whether the sample is at a defocused position based on a translational shift between a first wavelength channel and a second wavelength channel of the captured digital image; in response to the determination that the sample is in the defocused position, determine a defocus distance based on the transitional shift; and move a z-stage of the bright field image to a focus position based on the determined defocus distance; wherein the computing system is further configured to: determine a cross-talk first wave length channel and a cross-talk second wavelength channel of the captured digital image of the sample illuminated at the first and second incident angles with a first and a second wavelength light simultaneously; determine a first wavelength channel of the captured image illuminated with the first wavelength light at the first incident angle; determine a second wavelength channel of the captured image illuminated with the second wavelength light at the second incident angle; determine cross-talk coefficients based on the cross-talk first wavelength channel, the cross-talk second wavelength channel, the first wavelength channel and the second wavelength channel; and determine a corrected first wavelength channel and a corrected second wavelength channel based on the cross-talk first wavelength channel, the second wavelength channel and the cross-talk coefficients; wherein in determining the transitional shift, the computing system is configured to: maximize mutual information between the corrected first wavelength channel and the corrected second wavelength channel based on gradient descent; and wherein the computing system is further configured to: generate an in focus high-resolution image of a plurality of tiles of the sample; and generate the whole slide image of the sample based on the in focus high-resolution images of the plurality of tiles of the sample.

2. The system in claim 1 , wherein the sample is illuminated from the first incident angle with a red light source and from the second incident angle with a green light source.

3. The system in claim 1 , wherein the sample is illuminated from the first incident angle with a red light source and from the second incident angle with a blue light source.

4. The system in claim 1 , wherein the sample is illuminated from the first incident angle with a green light source and from the second incident angle with a blue light source.

5. The system in claim 1 , wherein the sample is illuminated from the first incident angle with a green light source and from the second incident angle with a blue light source and from a third incident angle with a red light source.

6. The system in claim 1 , wherein the first wavelength light and/or the second wavelength light is at least one of a light emitting diode, a laser diode and a halogen lamp with a filter.

7. The system in claim 1 , wherein in determining the transitional shift, the computing system is configured to: plot a cross-correlation between the corrected first wavelength channel and the corrected second wavelength channel; and determine a maximum point of the cross-correlation plot.

8. A method to capture a whole slide image of a sample, the method comprising: capturing, via a camera, a bright field image of a sample; capturing, via the camera, a digital image of the sample illuminated from a first incident angle at a first wavelength and a second incident angle at a second wavelength; determining, via a computing system, whether the sample is at a defocused position based on a transitional shift between a first wavelength channel and a second wavelength channel of the captured digital image; in response to the determination that the sample is in the defocused position, determining, via the computing system, a defocus distance based on the transitional shift; and moving a z-stage of the bright field image to a position where the image is in focus based on the determined defocus distance; determining, via the computing system, a cross-talk first wave length channel and a cross-talk second wavelength channel of the captured digital image of the sample illuminated at the first and second incident angles with a first and a second wavelength light simultaneously; determining, via the computing system, a first wavelength channel of the captured image illuminated with the first wavelength light at the first incident angle; determining, via the computing system, a second wavelength channel of the captured image illuminated with the second wavelength light at the second incident angle; determining, via the computing system, cross-talk coefficients based on the cross-talk first wavelength channel, the cross-talk second wavelength channel, the first wavelength channel and the second wavelength channel; determining, via the computing system, a corrected first wavelength channel and a corrected second wavelength channel based on the cross-talk first wavelength channel, the second wavelength channel and the cross-talk coefficients; maximizing mutual information between the corrected first wavelength channel and the corrected second wavelength channel based on gradient descent; generating, via the computing system, an in focus high-resolution image of a plurality of tiles of the sample; and generating, via, the computing system, the whole slide image of the sample based on the in focus high-resolution images of the plurality of tiles of the sample.

9. The method in claim 8 , wherein the sample is illuminated from the first incident angle with a red light source and from the second incident angle with a green light source.

10. The method in claim 8 , wherein the sample is illuminated from the first incident angle with a red light source and from the second incident angle with a blue light source.

11. The method in claim 8 , wherein the sample is illuminated from the first incident angle with a green light source and from the second incident angle with a blue light source.

12. The method in claim 8 , wherein the sample is illuminated from the first incident angle with a green light source and from the second incident angle with a blue light source and from a third incident angle with a red light source.

13. The method in claim 8 , wherein the first wavelength and/or the second wavelength light source is at least one of a light emitting diode, a laser diode and a halogen lamp with a filter.

14. The method in claim 8 , further comprising, plotting, via the computing system, a cross-correlation between the corrected first wavelength channel and the corrected second wavelength channel; and determining, via the computing system, a maximum point of the cross-correlation plot.